Occupancy Variability Base

	Function		Occupancy Variability
	Description		This tool, which is relatively rich, permits a user to monitor how occupancy counts vary over time.
	Access		Workspace > KPIs > Occupancy Variability
	Input		TV		Identifier of an active Traffic Volume of which the reference location is an Airspace = mostly a sector (elemntary or collapsed)
			WEF		Start of the trending period (hh:mm or hhmm)
			UNT		End of the trending period (hh:mm or hhmm)
			Duration		The Occupancy Duration being the period during which the flight is considered to be in the airspace starting at the time of the count bar and ending duration later. So the counts at the end of the trending period include also flights until (UNT+Duration). Durations for which an OTMV is defined are indicated with *n* (n being a number from 1 to 20). When an OTMV is defined the sustain or peak values can be indicated to alert the observer.
			Proposals		The Occupancy Counts can be based on the 'normal' traffic which is the traffic as it was filed or it can be based on the 'proposal' traffic which includes the flights that are being considered for a STAM measure. If a flight is in a proposed STAM measure, then the counts will reflect its absence or presence in the sector due to a STAM. This allows what-if analysis of the effect of a STAM.
			Algorithm		[ DEFAULT | NM | INTERNAL_PSEUDO_NM | INTERNAL_PROBABILITIC ]
					This tool includes some experimentation on different traffic counting mechanisms. Therefore, numerous HMIs (and for example this Occupancy Variability) offer the possibility to choose an algorithm : • DEFAULT – allow the prototype to choose the best algorithm (normally NM) • NM – use the “real” NM traffic counts • INTERNAL_PSEUDO_NM – a local count engine used pre-R19   • INTERNAL_PROBABILISTIC – a local count engine that performs probabilistic counts. The probabilistic counts engine takes into account an estimation of the flight ETO imprecision. For flights still on the ground, the imprecision is relatively high. For airborne flights, the imprecision decreases as the flight approaches the TV in question.  These imprecisions are then taken into account when computing the counts.
			Auto Refresh Mode		never, every 30s, min, 2 min, 3 min, 4 min, 5 min
	Output		The tool will perform a B2B query, obtaining the occupancy counts for a defined time period of [WEF,UNT[  and then compute a number of KPIs. The main KPI is the hotspot variability which indicates for each polling the evolution of the occupancy expressed as a bar ranging from the lowest observed value to the highest observed value since the polling began. The current observed occupancy is indicated as a line. Other KPIs available in or from this graph are explained in separate pages.
					This tool, which is relatively rich, permits a user to monitor how occupancy counts vary over time. The initial screen shot shows very little data    • A mini-FMP monitor , displaying the occupancy monitor bar    • The occupancy counts, shown in a graph format.    • Some initial statistics concerning the nature of the traffic –       o Number of airborne flights       o Number of flights in the 1st 10 minutes of being airborne (i.e. SID/climb phase)       o Number of flights on the ground       o Number of flights  on the ground and suspended       o Number of flights on the ground at A-CDM airports        o Number of flights on the ground at A-CDM airports and not yet T-DPI_s pre-sequenced       o Number of flights with slots       o Number of flights pre-allocated slots (>2h EOBT) The user leaves the tool running, with the auto-refresh feature active. Then, every ‘x’ minutes, the tool will redo the same occupancy count query. The obtained results (called a new ‘observation’) are then combined with the previous observations. Firstly, the mini-FMP monitor now has several bars. They all relate, to the same traffic volume query, but correspond to the different “observation” times . This enables the user to see how the “hotness” of the TV evolves over time. Secondly, the occupancy counts graph displays the latest occupancy counts, along with vertical bars that indicate the min/max occupancy values that have been observed. Here we can concretely see the variability in the counts. Thirdly, the bubble graph gives a view on how flight entry times have evolved across the various observations. The horizontal axis shows “Time in past”, where zero represents “now” (or to be more precise, the last observation time). The vertical axis indicates the difference in TV entry time (early/late). The flights are then divided into categories –    • On time – arriving within a [-5, +5] margin of the original expectation.    • Late – arriving [+5, +15] minutes later than originally expected    • Very late – arrive [15, …] minutes later than originally expected    • Early – arriving [-15, -5] minutes earlier than originally expected    • Very early – arrive [ …, -15] minutes earlier than originally expected Note: the bubble graph can be displayed as columns by clicking on the “Show Columns” button. Clicking on a bubble/column will bring up a pie chart detailing the flight status distribution.